^iSK',i8gf m*}  2feme  Multiple  Citrates.  169 
It  would  though  be  a  very  difficult  matter  to  ascertain  the  identity  of 
the  products.  Furthermore  the  possibilities  of  transformation  by 
-even  moderate  heat  or  merely  time  at  ordinary  temperatures  are  so 
immanent  that  scarcely  any  certainty  can  prevail  in  this  regard.  The 
effects  of  heat  either  alone  or  in  conjunction  with  other  agents  in  the 
production  of  metamerism  are  predominant.  It  is  the  writer's  belief 
that  metameric  states  are  the  result  of  intramolecular  substitution 
primarily  engendered  and  sustained  by  heat. 
If  for  illustration  in  the  last  formula,  the  effect  of  heat  should  mo- 
mentarily weaken  a  ferric  radicle  either  citric  radicle  might  attract  the 
basic  residue  joined  with  the  phosphoric  residue.  This  action  would 
momentarily  strengthen  the  phosphoric  radicle  with  a  tendency  to  the 
Incidentlv  freer  iron.  Should  the  new  combination  be  firmer  as  re- 
gards  heat  a  more  permanent  compound  of  the  following  structure 
would  result : 
Fe=P04— Fe  (7) 
I  II 
Na2=Ci  Ci— NH4 
Whatever  the  intra-molecular  condition  may  be  this  much  is  certain 
that  the  salt  has  a  very  light  apple-green  tint,  a  sweet  acidine  taste, 
scales  finely  and  does  not  precipitate  elixir  of  cinchona. 
A  remarkably  small  amount  of  heat  often  suffices  to  induce  intra- 
molecular change  as  well  as  molecular  disintegration  of  chemical  com- 
pounds ordinarily  stable.  The  writer  has  elsewhere  stated  (Amer. 
Jour.  Phar.,  March  1883)  that  when  equal  equivalents  of  ferric  cit- 
rate and  sodium  bicarbonate  react  a  hydrocitrate  of  sodium  and  iron 
results.  But  when  two  equivalents  of  the  carbonate  is  added  only 
half  the  carbonic  anhydrate  is  expelled.  The  writer  has  since  found 
that  the  action  of  ammonium  bicarbonate  is  analogous.  Yet  when  the 
two  equivalent  carbonate  solution  is  evaporated  half  the  ammonia  and 
all  the  carbonic  anhydrate  is  dissipated  whilst  the  monammonic  hy- 
dro-citrate 
OH — Fe = Ci — NH4  (8) 
remains  in  elegant  red-brown  scales.  When  caustic  ammonia 
is  added  to  ferric  citrate  to  saturation  two  equivalents  of  the 
base  are  absorbed.  On  evaporating  the  solution  by  means  of  a  mod- 
erate heat  the  hydro-citrate  just  noted  only  remains.  These  results 
show  that  an  excess  of  free  volatile  base  or  its  carbonate  when  added 
to  ferric  citrate  suffer  total  expulsion  when  heated  to  dryness.    A  so- 
